- Email: [email protected]
Application of oxidized metallic surfaces as a medium to store biochemical agents with antimicrobial properties
Surface & Coatings Technology 372 (2019) 312–318
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Application of oxidized metallic surfaces as a medium to store biochemical agents with antimicrobial properties Héctor M. Espejo, David F. Bahr
T
⁎
School of Materials Engineering, Purdue University, 701 West Stadium Ave., West Lafayette, IN 47907, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Cracked oxide morphology Desorption Nisin
Creating air-stable antimicrobial metallic surfaces that are resistant to washing could impact both food handling and medical applications. This work analyzes the release kinetics of nisin nanoparticles (an antimicrobial peptide) from cracked and uncracked oxide films on stainless steel and titanium to develop an antibacterial metallic surface. The cracked morphology creates mudflat crack patterns that penetrate approximately 4 μm into the substrate, deeper than the 100 nm thick oxides. The desorption of the peptide from uncracked oxides on stainless steel is promoted by acidic pH, slows down at neutral, and stops at basic pH. Increases in temperature contribute to significant accelerated desorption process only at acidic pH. Cracked oxide films effectively immobilized nisin in the cracks. For Ti-6Al-4 V, even under the harshest conditions tested (50 °C and pH 2) the peptide did not release into solution after 24 h. Quantitative depth profiling detected nitrogen to depths of up to 2.8 μm, indicating the presence of nisin in the cracks. A simple test of the antibacterial performance of these surfaces against Listeria monocytogenes has been demonstrated.
1. Introduction Infectious diseases result from pathogenic microbes (bacteria, fungi, parasites, etc.) and kill more people worldwide than any other single cause [1]. Microbial growth needs to be controlled, especially in the medical and food industries. Operating rooms require a sterile environment because they are filled with a number of potentially infectious objects. [2]. Raw vegetables [3], dairy products [4] or meat [5] can get contaminated when they are in contact with cutting tables, knives [6], slicers [7], conveyor belts [8], and during transport. These surfaces need to be cleaned and sanitized often to avoid food contamination. Despite cleaning procedures in hospitals, an estimated 722,000 nosocomial (hospital-acquired) infections occurred in the USA in 2011, resulting in nearly 75,000 deaths [2]. Similarly, several outbreaks of pathogens in food have happened in the last several years in products like cantaloupes, ice cream, frozen vegetables and raw milk cheese [9]. Current sterilization and cleaning/sanitation routines are not always sufficient for microbial control. Antimicrobial agents are used to kill bacteria and other microorganisms. If these agents are to be effective antimicrobial coatings on solid surfaces, they may either preclude microbial attachment to material surfaces or kill microorganisms that come into contact with the surface (biocidal property). Some examples of antimicrobial agents that
⁎
have been used with this purpose include metallic nanoparticles [10,11], N-halamines [12,13], quaternary ammonium compounds [14,15], essential oils [16,17], and antimicrobial peptides (AMPs) [18,19]. AMPs have attracted attention as antimicrobial agents due to their natural origin, broad antimicrobial spectrum and their efficacy at very low concentrations. More than 3000 antimicrobial peptides have been discovered but at the current time only one of them, nisin, is legally approved in more than 50 countries to be used as a food preservative [20]. Nisin is effective against several Gram-positive bacteria (such as Listeria monocytogenes and Clostridium botulinum), and may also be effective against Gram-negative bacteria [21] when combined with chelators such as ethylenediaminetetraacetic acid (EDTA), heat treatment or freezing. The ability to immobilize (i.e. “store”) antimicrobial peptides, such as nisin, on food contact surfaces should confer antibacterial properties to the surface material. This includes attaching peptides on polymers for antimicrobial packaging films [22,23], and some studies have also been done on metals [24–27]. One of the big challenges to creating such coatings is to ensure stability; the AMP must physically remain active in the film/coating for as long as possible, independent of the environmental conditions the contact surface is exposed to, but must not be so tightly bound (i.e. chemically) that it will not provide antimicrobial
Corresponding author. E-mail address: [email protected] (D.F. Bahr).
https://doi.org/10.1016/j.surfcoat.2019.05.059 Received 23 January 2019; Received in revised form 29 April 2019; Accepted 20 May 2019 Available online 21 May 2019 0257-8972/ © 2019 Elsevier B.V. All rights reserved.
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
2.3. Nisin
performance. Several studies have examined nisin coating/adhesion to polymeric materials. Lee et al. [28] determined that the migration of nisin from coated paper stopped after eight days, and the maximum equilibrium concentration of nisin released into solution was in the range of 222–241 μg/ml. This concentration corresponded to 8.6–9.3% of the total nisin content incorporated in the coating layer. Rossi et al. [29] found that nisin release from whey protein isolate edible films was favored at lower pH. At pH 7 there was only a small release of nisin because the AMP is below its isoelectric point (pI close to 10), bearing a positive charge, which may interact with the negative whey protein charges (pI 4.2–5.3) in the film. At pH 4, nisin shows an electrostatic repulsion with film proteins, because both molecules have a net positive charge, which favors nisin release to aqueous solution. Chacko et al. [30] measured nisin release from a corn zein membrane and determined that the release decreased as the concentration of corn zein increased. Bastarrachea et al. [31] studied the release kinetics of nisin from biodegradable poly(butylene adipate-co-terephthalate) (PBAT) films into water and found that the concentration of nisin in solution at equilibrium was higher as the temperature increased, and that the diffusion rates from PBAT were higher if comparing with other materials used in food packaging. To the extent of our knowledge, there is no work employing microscopic surface topography modification to “store” an AMP (in this case where the topography is formed by pulsed laser oxidation to modify the surface). Thin film fractures have been systematically investigated as a method to generate micro- and nano-scale structures by subjecting them to high residual stresses during the fabrication process [32]. By using the appropriate parameters, lasers are able to create features at micro- and nano-scale that can be used to store antimicrobial agents or other compounds. The present work aims to demonstrate that nisin can be physically adsorbed and stored in artificially created cracks on 304 L stainless steel and Ti-6Al-4 V, as well as doing an analysis of the release kinetics of nisin from those cracks and the antibacterial performance of nisin even after immobilization on oxide layers with cracked morphology.
Ultrapure nisin Z (Handary SA, Belgium) was used to prepare a solution of 1.0 mg/ml in sterile deionized water. A very small volume of this solution (max. 2 drops) was placed, using a pipette, on the fabricated oxide coatings, trying to cover as much as possible of the 6 × 6 mm area of the oxide. Subsequently, the coupons were placed inside a desiccator connected to a vacuum pump, and vacuum was applied for 10 min with the purpose to infuse the nisin solution, aiming to immobilize the peptide on the surface of U-SS or into the cracks of CSS and Ti alloy. After vacuum treatment, coupons were immersed for 10 s in 3 ml of sterile deionized water with the purpose to remove the “coffee stain” left on the coating after applying vacuum. 2.4. Release tests Three different release solutions were prepared: One of deionized water (pH ≈ 6), one of HCl 0.01 M (pH ≈ 2) and one of NaOH 0.01 M (pH ≈ 10). Temperature was varied at two levels: room temperature (RT) and 50 °C, by using a mini-dry bath (Benchmark). Experiments were performed immersing each coupon in 4 ml of release solution and sampling at different intervals of time. A bicinchoninic acid (BCA) kit from G-Biosciences (St. Louis, MO) was used as colorimetric method to determine the amount of nisin. Bovine serum albumin was used as a standard to create a calibration curve, and the regression line for this curve was expressed as the following:
Y = 0.0017 X + 0.0204 (R2 = 0.9952)
(1)
where X (μg/ml) is the concentration of nisin, and Y is the absorbance. A volume of 0.3 ml of release solution was taken at each time interval and placed into refrigeration at 4 °C. Once all samples of release solution were collected, a volume of 0.1 ml (taken from the 0.3 ml previously sampled) at each time interval was vigorously mixed with 1 ml of the BCA kit's working reagent. Absorbance was measured at 562 nm, no more than 10 min after mixing the working reagent with the 0.1 ml of release solution, using a SpectraMax 384 Plus UV/vis spectrophotometer [31], to determine the nisin concentrations based on Eq. (1). All experiments were made in duplicate. The accumulative release percentage of nisin from the oxide films is [34,35]:
2. Experimental 2.1. Stainless steel
Accumulative release (%) =
Two austenitic 304 L stainless steel (henceforth SS) coupons (12.5 × 12.5 × 3.4 mm) were irradiated using an Er-doped, glass-fiber laser (λ = 1.54 μm) with the conditions described in [33]. One of these coupons was irradiated (6 × 6 mm area) at a scan speed of 500 mm/s; the second coupon was irradiated (6 × 6 mm area) using varied scan speeds (10, 50, 90 and 130 mm/s). Colored oxide layers appeared as a result of the laser irradiation and these layers are easily visible at naked eye. Scanning electron microscopy was used to confirm that the first coupon (irradiated at 500 mm/s) did not develop significant cracks (Fig. 1a), while the second (irradiated at lower scan speeds) developed “mudflat” type cracks randomly distributed all over the surface (Fig. 1b). Henceforth the coupons will be named U-SS (for uncracked oxide layer on stainless steel) and C-SS (for cracked oxide layer on stainless steel). The color of the oxide layer is related to the film thickness, which depends on the employed laser scan rate; in this case, optically, the films appear to have a goldish color for U-SS, and brown, greenish and pink for C-SS.
Mt x 100 M0
(2)
where Mt (μg) is the released nisin at time t, M0 (μg) is the estimation of the total trapped or entrapped nisin in the oxide layer (surface for U-SS and including into the cracks for C-SS). 2.5. Depth profile A compositional depth profile analysis was performed on the Ti alloy, using a glow discharge spectrometer GDS-850A (LECO Corp., Ohio, USA) with an RF 4 mm lamp, operating at 30 W and 1000 V, with an acquisition rate of 100 s−1 and profile duration of 600 s. 2.6. Antibacterial performance With the purpose to demonstrate that nisin keeps its antibacterial properties even when immobilized on the cracked oxide layer of a metallic substrate, a test was performed against Listeria monocytogenes, a bacterium able to form biofilms and responsible for listeriosis in humans. The Ti alloy coupon was immersed for 48 h in a solution containing bacteria, followed by 48 h of incubation at 37 °C to promote the growth of attached bacteria. After incubation, the surface was washed to eliminate the non-attached bacteria, and then a cotton swab was used to scrub the oxide layer to transfer any bacteria to a growth media, where a standard plate count method was used to observe bacterial
2.2. Titanium alloy One Ti-6Al-4V (henceforth Ti alloy) coupon (12 × 12 × 3 mm) was irradiated using the same power conditions applied on SS, but with a fixed scan rate of 110 mm/s. A deep blue oxide layer appeared as a result of the laser irradiation and scanning electron microscopy revealed the presence of a pervasive network of cracks (Fig. 1c). 313
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 1. Overall surface topography as observed by scanning electron microscopy corresponding to: (a) U-SS, in which only some scratches and polishing lines are observed on the surface; (b) C-SS, in which cracks may be observed all over the surface and the areas with more cracks are marked in ovals; and (c) Ti alloy, showing a pervasive network of cracks as a result of laser irradiation.
was tested at room temperature and pH 6, while there was no measurable release when C-SS was subjected to the same conditions. This implies that for U-SS there was a slow detachment of nisin nanoparticles from the surface, while for C-SS the nisin is immobilized inside the cracks, taking much longer times to travel and reach the aqueous solution.
colonies. The test was done in duplicate for both the Ti alloy without nisin (negative control) and after coating with nisin. 3. Results and discussion 3.1. Characterization of the oxide layers
3.2.1. U-SS at room temperature and pH 6 The first sample of release solution was taken after 3 days of immersion and the BCA test indicated measureable nisin in solution. This test was run for 21 consecutive days. The relationship between time and accumulative nisin concentration is linear for this period of time (see Fig. 3), and the desorption process was still occurring when the test was stopped at three weeks.
3.1.1. Morphology Previous work [36–38] has determined that the chemistry and morphology of the U-SS (less than 100 nm thick) and the islands between cracks of C-SS and Ti alloy (both in the range 100–150 nm thick [39]) consist of a series of ridges and valleys, in which the vertical distance from the bottom of a valley to the top of a ridge is on the order of 3/5 of the total film thickness [37] Cross sections of oxide layers on C-SS and Ti alloy show both contain cracks with an approximate width of 100 nm at the surface and average depth of 4 μm (a relatively high aspect ratio) [39], which means these cracks penetrate through the 150 nm thick film and substantially into the substrate. The main difference is that cracks on C-SS are randomly distributed over the surface (Fig. 1b), while those on the Ti alloy form a clear interconnected network with approximately 10 μm of intercrack space [39] (Fig. 1c). The pervasive mudflat cracking of the Ti alloy indicates the oxides developed residual stresses during fabrication, likely due to differences in thermal expansion coefficients between oxide and substrate, and these stresses were partially released through cracking [36]. Previous analysis by nanoindentation has shown that these oxide layers are mechanically robust in regards to adhesion, and show no evidence of delamination under mechanical loading [33].
3.2.2. U-SS at room temperature and pH 10 After 21 consecutive days of immersion there was no apparent nisin release. This is likely due to the fact that nisin solubility is very low at basic pH, on the order of 0.25 mg/ml at pH 8.5 [42]. 3.2.3. U-SS at 50 °C and pH 2 The rise in temperature and decrease in pH definitely contributed to accelerate the desorption process from the oxide film. The relationship is still linear and, according to calculations, all the nisin retained on the oxide surface was released in less than 24 h. We estimated that the total release of nisin under these conditions took approximately 14 h (Fig. 4). The release rate (slope) shown on Fig. 4 is two orders of magnitude higher than that shown on Fig. 3;changing conditions to higher temperature and lower pH, causes the AMP release to occur approximately 100 times faster.
3.1.2. Composition The oxide layers on SS are mainly formed by a mixture of nonstoichiometric Fe, Cr and Mn oxides, however, for C-SS, it was detected an interlayer of Cr and Mn oxides (likely MnCr2O4) separating an outer layer (likely Fe3O4) from the substrate [37], forming what other authors have called a “duplex structure” [40,41]. This interlayer was not observed for U-SS. Regarding the Ti alloy, its outer layer is formed by a mixture of non-stoichiometric Ti oxides followed by an interlayer (mainly Ti6O) and then the substrate [36].
3.2.4. C-SS at room temperature and pH 6 This test was run for 18 consecutive days. There was no measurable nisin desorption in this period of time. At the conclusion of this test the C-SS coupon was immediately transferred to the solution at 50 °C and pH 2. 3.2.5. C-SS at 50 °C and pH 2 Exposure to these conditions led to significant corrosion of the substrate (Fig. 5) and the test was stopped. Nevertheless, the corrosion was accompanied by a strong change in coloration of the release solution, demonstrating that nisin was effectively immobilized in the cracks for the immersion in water at room temperature and pH 6, and was released once the surface of the SS underwent corrosion.
3.2. Release from stainless steel Release tests were performed by changing temperature and pH of the solution. A summary of the eight tests done on SS is shown in Fig. 2. A relatively slow release of nisin into solution was obtained when U-SS 314
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 2. Summary of the tests performed on the SS coupons, changing temperature and pH. Nisin release is mainly driven by pH, while rise in temperature contributes to accelerate the reaction.
Fig. 4. U-SS at 50 °C and pH 2. Approximately linear relationship between time and the accumulative nisin release (μg/ml). According to calculations, it would take approximately 14 h to release all the nisin retained on the oxide layer.
Fig. 3. U-SS at room temperature and pH 6. Linear relationship between time and the accumulative nisin release (μg/ml), but even at long times there is still desorption. According to calculations, only 34% of the total nisin retained by the oxide film has been released after 21 days, meaning that, if the linear behavior is maintained, all nisin would be released after 63 days of immersion under these conditions.
stronger bonding of nisin to the surface. FTIR spectra in Fig. 6 show that the signals indicative of the molecular groups present in nisin [43] (and not the oxide) are present on the oxidized Ti alloy surface (Fig. 6) and can be used to determine when nisin is present on the surface of the oxide. Once it was confirmed that nisin was attached to the Ti alloy oxide layer, the same immersion tests described in the procedures were unable to significantly remove any nisin from the surface. Therefore, it was concluded that nisin, even under harsh cleaning conditions, was not released into the solution and remains present on the surface of the Ti alloy's oxide (as well as within the cracks that penetrate the substrate). To demonstrate the extent of nisin adsorption, a compositional depth profile analysis using GDS was performed. Fig. 7 shows the results of the compositional profile at different depths in the Ti alloy. Given that nisin (C143H230N42O37S7) contains nitrogen in its chemical
The tests made using the C-SS coupon demonstrate that physisorption using vacuum is a good method to infuse nisin into cracks as long as the oxide layer is maintained at conditions close to room temperature and pH 6. There must be significant physisorption between the nisin nanoparticles and the stainless steel surface and crack walls that precludes the nisin desorption. Nonetheless, when the conditions turn harsh (50 °C and pH 2), then nisin desorption is promoted and all the peptide is released from the surface into solution.
3.3. Release from Ti alloy When the same method of physisorption of nisin was used on the oxidized Ti samples as done with SS, Ti exhibited a significantly 315
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 5. Microstructures corresponding to: (a) C-SS before nisin coating, in which cracks may be observed all over the surface, and (b) C-SS after nisin coating and immersion into solution at 50 °C and pH 2, the corroded surface shows remnants of the linear structure from the oxide growth but no evidence of cracks suggesting the total material removed is at least as deep as the cracked network. Fig. 6. FTIR spectra for Ti alloy oxide layer: A = nisin nanoparticles, B = pristine oxide layer, C = after coating with nisin ZP. The infrared signals around 2800–2900 cm−1 correspond to the CeH stretching bands and those around 1500–1600 cm−1 to amine bands, both organic groups present in nisin. This is evidence that nisin is attached to the surface. The signal at 880 cm−1 in B, indicative of the TieO bonding, has been obscured by the presence of nisin on the surface in C.
cracks and was only present on the surfaces of the “islands” in between the mudflat cracks, then evidence of the N signal would have been removed within 100 nm of the depth profile. According to Fig. 7, the presence of oxygen is understandable given that the analysis was performed on the oxide layer (oxygen diffused into this layer during the laser irradiation process, both forming a distinct oxide but also being present as a solid solution impurity under the oxide) and the percentage of oxygen, as expected, varies inversely related to depth. However, for nitrogen, the only explanation for the detection of this element is the presence of nisin trapped into the cracks. The depth profile analysis revealed the presence of nitrogen even at 2.8 μm deep, which seems reasonable given that the depth of these cracks is between 1 and 6 μm [39]. Therefore, nisin remains attached to the crack walls of the Ti alloy even if this is subjected to harsh conditions of temperature and acidity. Nisin exhibits more affinity to the cracked oxide layer on the Ti alloy compared to SS. 3.4. Antibacterial performance and implications for cleaning Fig. 8 shows the presence of bacterial colonies which grew when Listeria monocytogenes was exposed to the oxidized surface without nisin, but there was no bacterial growth for the surface on which nisin was immobilized. This promising preliminary bacterial test will be followed in the future by more detailed tests of antibacterial performance of nisin-coated surfaces. These initial results are promising for the potential application of this technology in the cleaning and disinfection of reusable surgical/ medical devices, surfaces of medical instruments, and metallic surfaces
Fig. 7. Compositional depth profile performed on the Ti alloy. The presence of nitrogen until 2.8 μm deep strongly suggests that nisin is still attached in the cracks even if the oxide layer was exposed to harsh conditions (50 °C and pH 2).
structure and no other significant presence of nitrogen is found in the Ti alloy, the presence of this element was selected to provide evidence of nisin in the cracks of the oxide layer. If nisin had not penetrated the 316
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 8. Antibacterial performance of the oxide layer on Ti alloy: (a) Non-coated surface, displaying bacterial growth, (b) Coated surface with nisin, with no bacterial growth, demonstrating that this peptide keeps antibacterial properties when immobilized on the cracked morphology.
system of cracks created by nanosecond pulsed laser oxidation for both stainless steel and Ti-6Al-4V; nisin is more immobilized on the Ti surfaces than on the stainless steels. Immersion in neutral or basic pH's solutions, at both room and elevated temperatures, does not significantly remove nisin from the stored condition, but low pH encourages the nisin desorption. Initial tests show that nisin displays antibacterial properties against Listeria monocytogenes when immobilized on the cracked oxide layer of Ti alloy, but more work is required to fully quantify this effect.
and tools in the food industry. In practice, substances like hypochlorites, peroxyacetic acid (PAA), hydrogen peroxide, quaternary ammonium compounds (QAC's) and iodophors [44,45] are used as active ingredients in the formulation of disinfectants. Most of these substances have an alkaline nature and are effective on a broad pH range, suggesting that nisin immobilized on the surfaces would be viable after conventional washing treatments. Some notable exceptions where the nisin treatment may be adversely impacted by cleaning are when using iodophors, which are effective at pH levels between 2 and 5 [46]. Use of iodophors on nisin-coated surface would contribute to the desorption of the peptide, having a detrimental effect on surface functionality. In addition, enzymatic-based commercial cleaning products (amylase, lipase, etc.) may provide proteases that are capable of degrading nisin [47,48], as they do in the human body when ingesting food products containing nisin as preservative. The use of disinfectants containing proteases could create unintended reactions that affect the antibacterial performance of the surface. However, since these disinfectants are diluted in water before application (usually 1:10 or 1:100 v/v) [49], this could reduce the detrimental effect that acidic or enzymatic-based disinfectants could have on the antibacterial performance of nisin-coated oxide layers.
Acknowledgments We thank Susana Díaz-Amaya of the School of Materials Engineering at Purdue University for her assistance and guidance in performing the microbiological tests against Listeria, and Dr. Amanda Deering for access to the lab facilities for said testing in the Food Science Department at Purdue University. References [1] K. Huang, C. Yang, S. Huang, C. Chen, Y. Lu, Y. Lin, Recent advances in antimicrobial polymers: a mini-review, Int. J. Mol. Sci. 17 (1578) (2016) 1–14. [2] M. Cloutier, D. Mantovani, F. Rosei, Antibacterial coatings: challenges, perspectives, and opportunities, Trends Biotechnol. 33 (11) (2015) 637–652. [3] Y. Fu, A. Deering, A. Bhunia, Y. Yao, Pathogen biofilm formation on cantaloupe surface and its impact on the antibacterial effect of lauroyl arginate ethyl, Food Microbiol. 64 (2017) 139–144. [4] H. Ksontini, F. Kachouri, M. Hamdi, Dairy biofilm: impact of microbial community on raw milk quality, J. Food Qual. 36 (4) (2013) 282–290. [5] J. Wang, Characteristics and Practices that May Lead to Listeria monocytogenes Contamination in Retail Delis, PhD thesis Purdue University, 2014. [6] E. van Asselt, A. de Jong, R. de Jonge, M. Nauta, Cross-contamination in the kitchen: estimation of transfer rates for cutting boards, hands and knives, J. Appl. Microbiol. 105 (5) (2008) 1392–1401. [7] D. Chen, T. Zhao, M. Doyle, Transfer of foodborne pathogens during mechanical slicing and their inactivation by levulinic acid-based sanitizer on slicers, Food Microbiol. 38 (2014) 263–269. [8] R. Scheffler, Preventive measures to reduce the risk of cross contamination on direct food contact surfaces of conveyor belts, J. Hyg. Eng. Des. 5 (2013) 3–5. [9] Centers for Disease Control and Prevention, [Online]. Available: https://www.cdc. gov/foodsafety/outbreaks/index.html , Accessed date: 2 June 2018. [10] S. Shankar, A. Oun, J. Rhim, Preparation of antimicrobial hybrid nano-materials using regenerated cellulose and metallic nanoparticles, Int. J. Biol. Macromol. 107 (2018) 17–27 Part A. [11] A. Roy, M. Joshi, B. Butola, S. Malhotra, Antimicrobial and toxicological behavior of montmorillonite immobilized metal nanoparticles, Mater. Sci. Eng. C 93 (2018) 704–715. [12] Y. Wang, M. Yin, X. Lin, L. Li, Y. Sun, Tailored synthesis of polymer-brush-grafted
4. Conclusions This study has examined the stability of physisorbed nisin on the oxidized surfaces of commercial stainless steel and titanium. Nisin desorption from an oxidized 304 L stainless steel surface is mainly controlled by changes in pH. Acidic pH encourages nisin desorption, while neutral and basic pH slows down the desporption or even stops it. Increases in temperature, when combined with a low pH, accelerate the desorption process, but does not seem to be a contributing factor by itself. The presence of ≈100 nm wide, ≈4 μm deep surface cracks leads to a morphology that enables nisin to be effectively immobilized within the cracks. This cracked surface did not liberate any nisin even when immersed in water at room temperature and pH 6 for 18 days, however, as soon as it was subjected to 50 °C and pH 2, the oxide layer underwent corrosion and all nisin was rapidly released. The adhesion of nisin on cracked Ti surfaces is greater than on the oxidized stainless steels. Compositional depth profiling by GDS revealed that nisin remains attached to the walls of the cracks even after exposure to harsh simulated cleaning conditions. Nisin can be effectively stored inside the surface topography of a 317
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
[13]
[14]
[15]
[16]
[17]
[18] [19]
[20] [21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
Inc. [31] L. Bastarrachea, S. Dhawan, S. Sablani, J. Powers, Release kinetics of nisin from biodegradable poly(butylene adipate-co-terephthalate) films into water, J. Food Eng. 100 (2010) 93–101. [32] T. Cao, A. Wang, X. Liang, H. Tang, G. Auner, S. O'Salley, S. Ng, Patterned immobilization of antibodies in mechanically induced cracks, J. Phys. Chem. B 112 (2008) 2727–2733. [33] S. Lawrence, D. Adams, D. Bahr, N. Moody, Mechanical and electromechanical behavior of oxide coatings grown on stainless steel 304L by nanosecond pulsed laser irradiation, Surf. Coat. Technol. 235 (2013) 860–866. [34] H. Wang, R. Zhang, H. Zhang, S. Jiang, H. Liu, M. Sun, S. Jiang, Kinetics and functional effectiveness of nisin loaded antimicrobial packaging film based on chitosan/poly(vinyl alcohol), Carbohydr. Polym. 127 (2015) 64–71. [35] H. Wang, H. Liu, C. Chu, Y. She, S. Jiang, L. Zhai, S. Jiang, X. Li, Diffusion and antibacterial properties of nisin-loaded chitosan/poly(L-lactic acid) towards development of active food packaging film, Food Bioprocess Technol. 8 (2015) 1657–1667. [36] S. Lawrence, D. Adams, D. Bahr, N. Moody, Deformation and fracture of a mudflatcracked laser-fabricated, J. Mater. Sci. 48 (2013) 4050–4058. [37] S. Lawrence, D. Adams, D. Bahr, N. Moody, Environmental resistance of oxide tags fabricated on 304L stainless steel, Surf. Coat. Technol. 285 (2016) 87–97. [38] S. Lawrence, D. Adams, D. Bahr, N. Moody, Mechanical and electromechanical behavior of oxide coatings grown on stainless steel 304L by nanosecond pulsed laser irradiation, Surf. Coat. Technol. 235 (2013) 860–866. [39] H. Espejo, D. Bahr, Substrate cracking in Ti-6Al-4V driven by pulsed laser irradiation, Surf. Coat. Technol. 322 (2017) 46–50. [40] D. Adams, V. Hodges, D. Hirschfeld, M. Rodriguez, J. McDonald, P. Kotula, Nanosecond pulsed laser irradiation of stainless steel 304L: oxide growth and effects, Surf. Coat. Technol. 222 (2013) 1–8. [41] Z. Li, H. Zheng, K. Teh, Y. Liu, G. Lim, H. Seng, N. Yakovlev, Analysis of oxide formation induced by UV laser coloration of stainless steel, Appl. Surf. Sci. 256 (5) (2009) 1582–1588. [42] W. Liu, J. Hansen, Some chemical and physical properties of nisin, a small-protein antibiotic produced by Lactococcus lactis, Appl. Environ. Microbiol. 56 (1990) 2551–2558. [43] L. Coli, V. Todaro, P. Cunha, L. Rodrigues, F. Almada, C. Marques, H. Keiko, H. Castro, C. Rangel, V. Pereira, L. Mendes, Development and characterization of nisin nanoparticles as potential alternative for the recurrent vaginal candidiasis treatment, AAPS PharmSciTech 17 (6) (2016) 1421–1427. [44] M. Iñiguez, M. Avila, E. Iñiguez, P. Guerrero, M. Gutierrez, Antimicrobial activity of disinfectants commonly used in the food industry in Mexico, J. Glob. Antimicrob. Resist. 10 (2017) 143–147. [45] E. Gonzalez, P. Nandy, A. Lucas, V. Hitchins, Ability of cleaning-disinfecting wipes to remove bacteria from medical device surfaces, Am. J. Infect. Control 43 (2015) 1331–1335. [46] G. McDonnell, Antisepsis, Disinfection, and Sterilization: Types, Action, and Resistance, American Society for Microbiology, 2007. [47] V. Suganthi, E. Selvarajan, C. Subathradevi, V. Mohanasrinivasan, Lantibiotic nisin: natural preservative from Lactococcus lactis, Int. Res. J. Pharm. 3 (1) (2012) 13–19. [48] K. Muller-Auffermann, F. Grijalva, F. Jacob, M. Hutzler, Nisin and its usage in breweries: a review and discussion, J. Inst. Brew. 121 (2015) 309–319. [49] Centers for Disease Control and Prevention, [Online]. Available https://www.cdc. gov/infectioncontrol/guidelines/disinfection/disinfection-methods/chemical.html , Accessed date: April 2019.
mesoporous silicas with N-halamine and quaternary ammonium groups for antimicrobial applications, J. Colloid Interface Sci. 533 (2019) 604–611. J. Lin, F. Jiang, J. Wen, W. Lv, Y. Sun, Fluorinated and un-fluorinated N-halamines as antimicrobial and biofilm-controlling additives for polymers, Polymer 68 (2015) 92–100. E. Poverenov, K. Miri, Formation of contact active antimicrobial surfaces by covalent grafting of quaternary ammonium compounds, Colloids Surf. B: Biointerfaces 169 (2018) 195–205. N. Gunther, A. Abdul-Wakeel, E. Reichenberger, S. Al-Khalifa, K. Minbiole, Quaternary ammonium compounds with multiple cationic moieties (multiQACs) provide antimicrobial activity against Campylobacter jejuni, Food Control 94 (2018) 187–194. M. Bravo, G. Preston, R. Van der Hoorn, H. Townley, I. Thompson, Species-specific antimicrobial activity of essential oils and enhancement by encapsulation in mesoporous silica nanoparticles, Ind. Crop. Prod. 122 (2018) 582–590. N. Khorshidian, M. Yousefi, E. Khanniri, A. Mohammad, Potential application of essential oils as antimicrobial preservatives in cheese, Innovative Food Sci. Emerg. Technol. 45 (2018) 62–72. T. Ren, Y. Wang, Q. Yu, M. Li, Synthesis of antimicrobial peptide-grafted graphene oxide nanosheets with high antimicrobial efficacy, Mater. Lett. 235 (2019) 42–45. D. Holdbrook, S. Singh, Y. Choong, J. Petrlova, R. Saravanan, Influence of pH on the activity of thrombin-derived antimicrobial peptides, Biochim. Biophys. Acta Biomembr. 1860 (11) (2018) 2374–2384. L. Juncioni, A. Faustino, P. Gava, T. Vessoni, Nisin biotechnological production and application: a review, Trends Food Sci. Technol. 20 (2009) 146–154. A. Gharsallaoui, N. Oulahal, C. Joly, P. Degraeve, Nisin as a food preservative: part 1: physicochemical properties, antimicrobial activity, and main uses, Crit. Rev. Food Sci. Nutr. 56 (2016) 1262–1274. S. Singh, I. Park, Y. Shin, M. Lee, Y. Lee, Antimicrobial seafood packaging: a review, J. Food Sci. Technol. 53 (6) (2016) 2505–2518. Y. Hung, L. McLandsborough, J. Goddard, L. Bastarrachea, Antimicrobial polymer coatings with efficacy against pathogenic and spoilage microorganisms, LWT 97 (2018) 546–554. H. Yazici, G. Habib, K. Boone, M. Urgen, C. Tamerler, Self-assembling antimicrobial peptides on nanotubular titanium surfaces coated with calcium phosphate for local therapy, Mater. Sci. Eng. C 94 (2019) 333–343. K. Doiron, L. Beaulieu, R. St-Louis, K. Lemarchand, Reduction of bacterial biofilm formation using marine natural antimicrobial peptides, Colloids Surf. B: Biointerfaces 167 (2018) 524–530. A. Héquet, V. Humblot, J. Berjeaud, C. Pradier, Optimized grafting of antimicrobial peptides on stainless steel surface and biofilm resistance tests, Colloids Surf. B: Biointerfaces 84 (2011) 301–309. E. Parfenov, L. Parfenova, G. Dyakonov, K. Danilko, V. Mukaeva, R. Farrakhov, E. Lukina, R. Valiev, Surface functionalization via PEO coating and RGD peptide for nanostructured titanium implants and their in vitro assessment, Surf. Coat. Technol. 357 (2019) 669–683. C. Lee, D. An, H. Park, D. Lee, Wide-spectrum antimicrobial packaging materials incorporating nisin and chitosan in the coating, Packag. Technol. Sci. 16 (3) (2003) 99–106. G. Rossi, J. Han, B. Garcia, E. Castaño, C. Regalado, Effect of temperature, pH and film thickness on nisin release from antimicrobial whey protein isolate edible films, J. Sci. Food Agric. 89 (2009) 2492–2497. J. Chacko, M. Lalpuria, J. Floros and R. Anantheswaran, "Controlled Release of Nisin from Biopolymer Films," The Pennsylvania State University - General Mills
318
Contents lists available at ScienceDirect
Surface & Coatings Technology journal homepage: www.elsevier.com/locate/surfcoat
Application of oxidized metallic surfaces as a medium to store biochemical agents with antimicrobial properties Héctor M. Espejo, David F. Bahr
T
⁎
School of Materials Engineering, Purdue University, 701 West Stadium Ave., West Lafayette, IN 47907, USA
A R T I C LE I N FO
A B S T R A C T
Keywords: Cracked oxide morphology Desorption Nisin
Creating air-stable antimicrobial metallic surfaces that are resistant to washing could impact both food handling and medical applications. This work analyzes the release kinetics of nisin nanoparticles (an antimicrobial peptide) from cracked and uncracked oxide films on stainless steel and titanium to develop an antibacterial metallic surface. The cracked morphology creates mudflat crack patterns that penetrate approximately 4 μm into the substrate, deeper than the 100 nm thick oxides. The desorption of the peptide from uncracked oxides on stainless steel is promoted by acidic pH, slows down at neutral, and stops at basic pH. Increases in temperature contribute to significant accelerated desorption process only at acidic pH. Cracked oxide films effectively immobilized nisin in the cracks. For Ti-6Al-4 V, even under the harshest conditions tested (50 °C and pH 2) the peptide did not release into solution after 24 h. Quantitative depth profiling detected nitrogen to depths of up to 2.8 μm, indicating the presence of nisin in the cracks. A simple test of the antibacterial performance of these surfaces against Listeria monocytogenes has been demonstrated.
1. Introduction Infectious diseases result from pathogenic microbes (bacteria, fungi, parasites, etc.) and kill more people worldwide than any other single cause [1]. Microbial growth needs to be controlled, especially in the medical and food industries. Operating rooms require a sterile environment because they are filled with a number of potentially infectious objects. [2]. Raw vegetables [3], dairy products [4] or meat [5] can get contaminated when they are in contact with cutting tables, knives [6], slicers [7], conveyor belts [8], and during transport. These surfaces need to be cleaned and sanitized often to avoid food contamination. Despite cleaning procedures in hospitals, an estimated 722,000 nosocomial (hospital-acquired) infections occurred in the USA in 2011, resulting in nearly 75,000 deaths [2]. Similarly, several outbreaks of pathogens in food have happened in the last several years in products like cantaloupes, ice cream, frozen vegetables and raw milk cheese [9]. Current sterilization and cleaning/sanitation routines are not always sufficient for microbial control. Antimicrobial agents are used to kill bacteria and other microorganisms. If these agents are to be effective antimicrobial coatings on solid surfaces, they may either preclude microbial attachment to material surfaces or kill microorganisms that come into contact with the surface (biocidal property). Some examples of antimicrobial agents that
⁎
have been used with this purpose include metallic nanoparticles [10,11], N-halamines [12,13], quaternary ammonium compounds [14,15], essential oils [16,17], and antimicrobial peptides (AMPs) [18,19]. AMPs have attracted attention as antimicrobial agents due to their natural origin, broad antimicrobial spectrum and their efficacy at very low concentrations. More than 3000 antimicrobial peptides have been discovered but at the current time only one of them, nisin, is legally approved in more than 50 countries to be used as a food preservative [20]. Nisin is effective against several Gram-positive bacteria (such as Listeria monocytogenes and Clostridium botulinum), and may also be effective against Gram-negative bacteria [21] when combined with chelators such as ethylenediaminetetraacetic acid (EDTA), heat treatment or freezing. The ability to immobilize (i.e. “store”) antimicrobial peptides, such as nisin, on food contact surfaces should confer antibacterial properties to the surface material. This includes attaching peptides on polymers for antimicrobial packaging films [22,23], and some studies have also been done on metals [24–27]. One of the big challenges to creating such coatings is to ensure stability; the AMP must physically remain active in the film/coating for as long as possible, independent of the environmental conditions the contact surface is exposed to, but must not be so tightly bound (i.e. chemically) that it will not provide antimicrobial
Corresponding author. E-mail address: [email protected] (D.F. Bahr).
https://doi.org/10.1016/j.surfcoat.2019.05.059 Received 23 January 2019; Received in revised form 29 April 2019; Accepted 20 May 2019 Available online 21 May 2019 0257-8972/ © 2019 Elsevier B.V. All rights reserved.
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
2.3. Nisin
performance. Several studies have examined nisin coating/adhesion to polymeric materials. Lee et al. [28] determined that the migration of nisin from coated paper stopped after eight days, and the maximum equilibrium concentration of nisin released into solution was in the range of 222–241 μg/ml. This concentration corresponded to 8.6–9.3% of the total nisin content incorporated in the coating layer. Rossi et al. [29] found that nisin release from whey protein isolate edible films was favored at lower pH. At pH 7 there was only a small release of nisin because the AMP is below its isoelectric point (pI close to 10), bearing a positive charge, which may interact with the negative whey protein charges (pI 4.2–5.3) in the film. At pH 4, nisin shows an electrostatic repulsion with film proteins, because both molecules have a net positive charge, which favors nisin release to aqueous solution. Chacko et al. [30] measured nisin release from a corn zein membrane and determined that the release decreased as the concentration of corn zein increased. Bastarrachea et al. [31] studied the release kinetics of nisin from biodegradable poly(butylene adipate-co-terephthalate) (PBAT) films into water and found that the concentration of nisin in solution at equilibrium was higher as the temperature increased, and that the diffusion rates from PBAT were higher if comparing with other materials used in food packaging. To the extent of our knowledge, there is no work employing microscopic surface topography modification to “store” an AMP (in this case where the topography is formed by pulsed laser oxidation to modify the surface). Thin film fractures have been systematically investigated as a method to generate micro- and nano-scale structures by subjecting them to high residual stresses during the fabrication process [32]. By using the appropriate parameters, lasers are able to create features at micro- and nano-scale that can be used to store antimicrobial agents or other compounds. The present work aims to demonstrate that nisin can be physically adsorbed and stored in artificially created cracks on 304 L stainless steel and Ti-6Al-4 V, as well as doing an analysis of the release kinetics of nisin from those cracks and the antibacterial performance of nisin even after immobilization on oxide layers with cracked morphology.
Ultrapure nisin Z (Handary SA, Belgium) was used to prepare a solution of 1.0 mg/ml in sterile deionized water. A very small volume of this solution (max. 2 drops) was placed, using a pipette, on the fabricated oxide coatings, trying to cover as much as possible of the 6 × 6 mm area of the oxide. Subsequently, the coupons were placed inside a desiccator connected to a vacuum pump, and vacuum was applied for 10 min with the purpose to infuse the nisin solution, aiming to immobilize the peptide on the surface of U-SS or into the cracks of CSS and Ti alloy. After vacuum treatment, coupons were immersed for 10 s in 3 ml of sterile deionized water with the purpose to remove the “coffee stain” left on the coating after applying vacuum. 2.4. Release tests Three different release solutions were prepared: One of deionized water (pH ≈ 6), one of HCl 0.01 M (pH ≈ 2) and one of NaOH 0.01 M (pH ≈ 10). Temperature was varied at two levels: room temperature (RT) and 50 °C, by using a mini-dry bath (Benchmark). Experiments were performed immersing each coupon in 4 ml of release solution and sampling at different intervals of time. A bicinchoninic acid (BCA) kit from G-Biosciences (St. Louis, MO) was used as colorimetric method to determine the amount of nisin. Bovine serum albumin was used as a standard to create a calibration curve, and the regression line for this curve was expressed as the following:
Y = 0.0017 X + 0.0204 (R2 = 0.9952)
(1)
where X (μg/ml) is the concentration of nisin, and Y is the absorbance. A volume of 0.3 ml of release solution was taken at each time interval and placed into refrigeration at 4 °C. Once all samples of release solution were collected, a volume of 0.1 ml (taken from the 0.3 ml previously sampled) at each time interval was vigorously mixed with 1 ml of the BCA kit's working reagent. Absorbance was measured at 562 nm, no more than 10 min after mixing the working reagent with the 0.1 ml of release solution, using a SpectraMax 384 Plus UV/vis spectrophotometer [31], to determine the nisin concentrations based on Eq. (1). All experiments were made in duplicate. The accumulative release percentage of nisin from the oxide films is [34,35]:
2. Experimental 2.1. Stainless steel
Accumulative release (%) =
Two austenitic 304 L stainless steel (henceforth SS) coupons (12.5 × 12.5 × 3.4 mm) were irradiated using an Er-doped, glass-fiber laser (λ = 1.54 μm) with the conditions described in [33]. One of these coupons was irradiated (6 × 6 mm area) at a scan speed of 500 mm/s; the second coupon was irradiated (6 × 6 mm area) using varied scan speeds (10, 50, 90 and 130 mm/s). Colored oxide layers appeared as a result of the laser irradiation and these layers are easily visible at naked eye. Scanning electron microscopy was used to confirm that the first coupon (irradiated at 500 mm/s) did not develop significant cracks (Fig. 1a), while the second (irradiated at lower scan speeds) developed “mudflat” type cracks randomly distributed all over the surface (Fig. 1b). Henceforth the coupons will be named U-SS (for uncracked oxide layer on stainless steel) and C-SS (for cracked oxide layer on stainless steel). The color of the oxide layer is related to the film thickness, which depends on the employed laser scan rate; in this case, optically, the films appear to have a goldish color for U-SS, and brown, greenish and pink for C-SS.
Mt x 100 M0
(2)
where Mt (μg) is the released nisin at time t, M0 (μg) is the estimation of the total trapped or entrapped nisin in the oxide layer (surface for U-SS and including into the cracks for C-SS). 2.5. Depth profile A compositional depth profile analysis was performed on the Ti alloy, using a glow discharge spectrometer GDS-850A (LECO Corp., Ohio, USA) with an RF 4 mm lamp, operating at 30 W and 1000 V, with an acquisition rate of 100 s−1 and profile duration of 600 s. 2.6. Antibacterial performance With the purpose to demonstrate that nisin keeps its antibacterial properties even when immobilized on the cracked oxide layer of a metallic substrate, a test was performed against Listeria monocytogenes, a bacterium able to form biofilms and responsible for listeriosis in humans. The Ti alloy coupon was immersed for 48 h in a solution containing bacteria, followed by 48 h of incubation at 37 °C to promote the growth of attached bacteria. After incubation, the surface was washed to eliminate the non-attached bacteria, and then a cotton swab was used to scrub the oxide layer to transfer any bacteria to a growth media, where a standard plate count method was used to observe bacterial
2.2. Titanium alloy One Ti-6Al-4V (henceforth Ti alloy) coupon (12 × 12 × 3 mm) was irradiated using the same power conditions applied on SS, but with a fixed scan rate of 110 mm/s. A deep blue oxide layer appeared as a result of the laser irradiation and scanning electron microscopy revealed the presence of a pervasive network of cracks (Fig. 1c). 313
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 1. Overall surface topography as observed by scanning electron microscopy corresponding to: (a) U-SS, in which only some scratches and polishing lines are observed on the surface; (b) C-SS, in which cracks may be observed all over the surface and the areas with more cracks are marked in ovals; and (c) Ti alloy, showing a pervasive network of cracks as a result of laser irradiation.
was tested at room temperature and pH 6, while there was no measurable release when C-SS was subjected to the same conditions. This implies that for U-SS there was a slow detachment of nisin nanoparticles from the surface, while for C-SS the nisin is immobilized inside the cracks, taking much longer times to travel and reach the aqueous solution.
colonies. The test was done in duplicate for both the Ti alloy without nisin (negative control) and after coating with nisin. 3. Results and discussion 3.1. Characterization of the oxide layers
3.2.1. U-SS at room temperature and pH 6 The first sample of release solution was taken after 3 days of immersion and the BCA test indicated measureable nisin in solution. This test was run for 21 consecutive days. The relationship between time and accumulative nisin concentration is linear for this period of time (see Fig. 3), and the desorption process was still occurring when the test was stopped at three weeks.
3.1.1. Morphology Previous work [36–38] has determined that the chemistry and morphology of the U-SS (less than 100 nm thick) and the islands between cracks of C-SS and Ti alloy (both in the range 100–150 nm thick [39]) consist of a series of ridges and valleys, in which the vertical distance from the bottom of a valley to the top of a ridge is on the order of 3/5 of the total film thickness [37] Cross sections of oxide layers on C-SS and Ti alloy show both contain cracks with an approximate width of 100 nm at the surface and average depth of 4 μm (a relatively high aspect ratio) [39], which means these cracks penetrate through the 150 nm thick film and substantially into the substrate. The main difference is that cracks on C-SS are randomly distributed over the surface (Fig. 1b), while those on the Ti alloy form a clear interconnected network with approximately 10 μm of intercrack space [39] (Fig. 1c). The pervasive mudflat cracking of the Ti alloy indicates the oxides developed residual stresses during fabrication, likely due to differences in thermal expansion coefficients between oxide and substrate, and these stresses were partially released through cracking [36]. Previous analysis by nanoindentation has shown that these oxide layers are mechanically robust in regards to adhesion, and show no evidence of delamination under mechanical loading [33].
3.2.2. U-SS at room temperature and pH 10 After 21 consecutive days of immersion there was no apparent nisin release. This is likely due to the fact that nisin solubility is very low at basic pH, on the order of 0.25 mg/ml at pH 8.5 [42]. 3.2.3. U-SS at 50 °C and pH 2 The rise in temperature and decrease in pH definitely contributed to accelerate the desorption process from the oxide film. The relationship is still linear and, according to calculations, all the nisin retained on the oxide surface was released in less than 24 h. We estimated that the total release of nisin under these conditions took approximately 14 h (Fig. 4). The release rate (slope) shown on Fig. 4 is two orders of magnitude higher than that shown on Fig. 3;changing conditions to higher temperature and lower pH, causes the AMP release to occur approximately 100 times faster.
3.1.2. Composition The oxide layers on SS are mainly formed by a mixture of nonstoichiometric Fe, Cr and Mn oxides, however, for C-SS, it was detected an interlayer of Cr and Mn oxides (likely MnCr2O4) separating an outer layer (likely Fe3O4) from the substrate [37], forming what other authors have called a “duplex structure” [40,41]. This interlayer was not observed for U-SS. Regarding the Ti alloy, its outer layer is formed by a mixture of non-stoichiometric Ti oxides followed by an interlayer (mainly Ti6O) and then the substrate [36].
3.2.4. C-SS at room temperature and pH 6 This test was run for 18 consecutive days. There was no measurable nisin desorption in this period of time. At the conclusion of this test the C-SS coupon was immediately transferred to the solution at 50 °C and pH 2. 3.2.5. C-SS at 50 °C and pH 2 Exposure to these conditions led to significant corrosion of the substrate (Fig. 5) and the test was stopped. Nevertheless, the corrosion was accompanied by a strong change in coloration of the release solution, demonstrating that nisin was effectively immobilized in the cracks for the immersion in water at room temperature and pH 6, and was released once the surface of the SS underwent corrosion.
3.2. Release from stainless steel Release tests were performed by changing temperature and pH of the solution. A summary of the eight tests done on SS is shown in Fig. 2. A relatively slow release of nisin into solution was obtained when U-SS 314
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 2. Summary of the tests performed on the SS coupons, changing temperature and pH. Nisin release is mainly driven by pH, while rise in temperature contributes to accelerate the reaction.
Fig. 4. U-SS at 50 °C and pH 2. Approximately linear relationship between time and the accumulative nisin release (μg/ml). According to calculations, it would take approximately 14 h to release all the nisin retained on the oxide layer.
Fig. 3. U-SS at room temperature and pH 6. Linear relationship between time and the accumulative nisin release (μg/ml), but even at long times there is still desorption. According to calculations, only 34% of the total nisin retained by the oxide film has been released after 21 days, meaning that, if the linear behavior is maintained, all nisin would be released after 63 days of immersion under these conditions.
stronger bonding of nisin to the surface. FTIR spectra in Fig. 6 show that the signals indicative of the molecular groups present in nisin [43] (and not the oxide) are present on the oxidized Ti alloy surface (Fig. 6) and can be used to determine when nisin is present on the surface of the oxide. Once it was confirmed that nisin was attached to the Ti alloy oxide layer, the same immersion tests described in the procedures were unable to significantly remove any nisin from the surface. Therefore, it was concluded that nisin, even under harsh cleaning conditions, was not released into the solution and remains present on the surface of the Ti alloy's oxide (as well as within the cracks that penetrate the substrate). To demonstrate the extent of nisin adsorption, a compositional depth profile analysis using GDS was performed. Fig. 7 shows the results of the compositional profile at different depths in the Ti alloy. Given that nisin (C143H230N42O37S7) contains nitrogen in its chemical
The tests made using the C-SS coupon demonstrate that physisorption using vacuum is a good method to infuse nisin into cracks as long as the oxide layer is maintained at conditions close to room temperature and pH 6. There must be significant physisorption between the nisin nanoparticles and the stainless steel surface and crack walls that precludes the nisin desorption. Nonetheless, when the conditions turn harsh (50 °C and pH 2), then nisin desorption is promoted and all the peptide is released from the surface into solution.
3.3. Release from Ti alloy When the same method of physisorption of nisin was used on the oxidized Ti samples as done with SS, Ti exhibited a significantly 315
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 5. Microstructures corresponding to: (a) C-SS before nisin coating, in which cracks may be observed all over the surface, and (b) C-SS after nisin coating and immersion into solution at 50 °C and pH 2, the corroded surface shows remnants of the linear structure from the oxide growth but no evidence of cracks suggesting the total material removed is at least as deep as the cracked network. Fig. 6. FTIR spectra for Ti alloy oxide layer: A = nisin nanoparticles, B = pristine oxide layer, C = after coating with nisin ZP. The infrared signals around 2800–2900 cm−1 correspond to the CeH stretching bands and those around 1500–1600 cm−1 to amine bands, both organic groups present in nisin. This is evidence that nisin is attached to the surface. The signal at 880 cm−1 in B, indicative of the TieO bonding, has been obscured by the presence of nisin on the surface in C.
cracks and was only present on the surfaces of the “islands” in between the mudflat cracks, then evidence of the N signal would have been removed within 100 nm of the depth profile. According to Fig. 7, the presence of oxygen is understandable given that the analysis was performed on the oxide layer (oxygen diffused into this layer during the laser irradiation process, both forming a distinct oxide but also being present as a solid solution impurity under the oxide) and the percentage of oxygen, as expected, varies inversely related to depth. However, for nitrogen, the only explanation for the detection of this element is the presence of nisin trapped into the cracks. The depth profile analysis revealed the presence of nitrogen even at 2.8 μm deep, which seems reasonable given that the depth of these cracks is between 1 and 6 μm [39]. Therefore, nisin remains attached to the crack walls of the Ti alloy even if this is subjected to harsh conditions of temperature and acidity. Nisin exhibits more affinity to the cracked oxide layer on the Ti alloy compared to SS. 3.4. Antibacterial performance and implications for cleaning Fig. 8 shows the presence of bacterial colonies which grew when Listeria monocytogenes was exposed to the oxidized surface without nisin, but there was no bacterial growth for the surface on which nisin was immobilized. This promising preliminary bacterial test will be followed in the future by more detailed tests of antibacterial performance of nisin-coated surfaces. These initial results are promising for the potential application of this technology in the cleaning and disinfection of reusable surgical/ medical devices, surfaces of medical instruments, and metallic surfaces
Fig. 7. Compositional depth profile performed on the Ti alloy. The presence of nitrogen until 2.8 μm deep strongly suggests that nisin is still attached in the cracks even if the oxide layer was exposed to harsh conditions (50 °C and pH 2).
structure and no other significant presence of nitrogen is found in the Ti alloy, the presence of this element was selected to provide evidence of nisin in the cracks of the oxide layer. If nisin had not penetrated the 316
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
Fig. 8. Antibacterial performance of the oxide layer on Ti alloy: (a) Non-coated surface, displaying bacterial growth, (b) Coated surface with nisin, with no bacterial growth, demonstrating that this peptide keeps antibacterial properties when immobilized on the cracked morphology.
system of cracks created by nanosecond pulsed laser oxidation for both stainless steel and Ti-6Al-4V; nisin is more immobilized on the Ti surfaces than on the stainless steels. Immersion in neutral or basic pH's solutions, at both room and elevated temperatures, does not significantly remove nisin from the stored condition, but low pH encourages the nisin desorption. Initial tests show that nisin displays antibacterial properties against Listeria monocytogenes when immobilized on the cracked oxide layer of Ti alloy, but more work is required to fully quantify this effect.
and tools in the food industry. In practice, substances like hypochlorites, peroxyacetic acid (PAA), hydrogen peroxide, quaternary ammonium compounds (QAC's) and iodophors [44,45] are used as active ingredients in the formulation of disinfectants. Most of these substances have an alkaline nature and are effective on a broad pH range, suggesting that nisin immobilized on the surfaces would be viable after conventional washing treatments. Some notable exceptions where the nisin treatment may be adversely impacted by cleaning are when using iodophors, which are effective at pH levels between 2 and 5 [46]. Use of iodophors on nisin-coated surface would contribute to the desorption of the peptide, having a detrimental effect on surface functionality. In addition, enzymatic-based commercial cleaning products (amylase, lipase, etc.) may provide proteases that are capable of degrading nisin [47,48], as they do in the human body when ingesting food products containing nisin as preservative. The use of disinfectants containing proteases could create unintended reactions that affect the antibacterial performance of the surface. However, since these disinfectants are diluted in water before application (usually 1:10 or 1:100 v/v) [49], this could reduce the detrimental effect that acidic or enzymatic-based disinfectants could have on the antibacterial performance of nisin-coated oxide layers.
Acknowledgments We thank Susana Díaz-Amaya of the School of Materials Engineering at Purdue University for her assistance and guidance in performing the microbiological tests against Listeria, and Dr. Amanda Deering for access to the lab facilities for said testing in the Food Science Department at Purdue University. References [1] K. Huang, C. Yang, S. Huang, C. Chen, Y. Lu, Y. Lin, Recent advances in antimicrobial polymers: a mini-review, Int. J. Mol. Sci. 17 (1578) (2016) 1–14. [2] M. Cloutier, D. Mantovani, F. Rosei, Antibacterial coatings: challenges, perspectives, and opportunities, Trends Biotechnol. 33 (11) (2015) 637–652. [3] Y. Fu, A. Deering, A. Bhunia, Y. Yao, Pathogen biofilm formation on cantaloupe surface and its impact on the antibacterial effect of lauroyl arginate ethyl, Food Microbiol. 64 (2017) 139–144. [4] H. Ksontini, F. Kachouri, M. Hamdi, Dairy biofilm: impact of microbial community on raw milk quality, J. Food Qual. 36 (4) (2013) 282–290. [5] J. Wang, Characteristics and Practices that May Lead to Listeria monocytogenes Contamination in Retail Delis, PhD thesis Purdue University, 2014. [6] E. van Asselt, A. de Jong, R. de Jonge, M. Nauta, Cross-contamination in the kitchen: estimation of transfer rates for cutting boards, hands and knives, J. Appl. Microbiol. 105 (5) (2008) 1392–1401. [7] D. Chen, T. Zhao, M. Doyle, Transfer of foodborne pathogens during mechanical slicing and their inactivation by levulinic acid-based sanitizer on slicers, Food Microbiol. 38 (2014) 263–269. [8] R. Scheffler, Preventive measures to reduce the risk of cross contamination on direct food contact surfaces of conveyor belts, J. Hyg. Eng. Des. 5 (2013) 3–5. [9] Centers for Disease Control and Prevention, [Online]. Available: https://www.cdc. gov/foodsafety/outbreaks/index.html , Accessed date: 2 June 2018. [10] S. Shankar, A. Oun, J. Rhim, Preparation of antimicrobial hybrid nano-materials using regenerated cellulose and metallic nanoparticles, Int. J. Biol. Macromol. 107 (2018) 17–27 Part A. [11] A. Roy, M. Joshi, B. Butola, S. Malhotra, Antimicrobial and toxicological behavior of montmorillonite immobilized metal nanoparticles, Mater. Sci. Eng. C 93 (2018) 704–715. [12] Y. Wang, M. Yin, X. Lin, L. Li, Y. Sun, Tailored synthesis of polymer-brush-grafted
4. Conclusions This study has examined the stability of physisorbed nisin on the oxidized surfaces of commercial stainless steel and titanium. Nisin desorption from an oxidized 304 L stainless steel surface is mainly controlled by changes in pH. Acidic pH encourages nisin desorption, while neutral and basic pH slows down the desporption or even stops it. Increases in temperature, when combined with a low pH, accelerate the desorption process, but does not seem to be a contributing factor by itself. The presence of ≈100 nm wide, ≈4 μm deep surface cracks leads to a morphology that enables nisin to be effectively immobilized within the cracks. This cracked surface did not liberate any nisin even when immersed in water at room temperature and pH 6 for 18 days, however, as soon as it was subjected to 50 °C and pH 2, the oxide layer underwent corrosion and all nisin was rapidly released. The adhesion of nisin on cracked Ti surfaces is greater than on the oxidized stainless steels. Compositional depth profiling by GDS revealed that nisin remains attached to the walls of the cracks even after exposure to harsh simulated cleaning conditions. Nisin can be effectively stored inside the surface topography of a 317
Surface & Coatings Technology 372 (2019) 312–318
H.M. Espejo and D.F. Bahr
[13]
[14]
[15]
[16]
[17]
[18] [19]
[20] [21]
[22] [23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
Inc. [31] L. Bastarrachea, S. Dhawan, S. Sablani, J. Powers, Release kinetics of nisin from biodegradable poly(butylene adipate-co-terephthalate) films into water, J. Food Eng. 100 (2010) 93–101. [32] T. Cao, A. Wang, X. Liang, H. Tang, G. Auner, S. O'Salley, S. Ng, Patterned immobilization of antibodies in mechanically induced cracks, J. Phys. Chem. B 112 (2008) 2727–2733. [33] S. Lawrence, D. Adams, D. Bahr, N. Moody, Mechanical and electromechanical behavior of oxide coatings grown on stainless steel 304L by nanosecond pulsed laser irradiation, Surf. Coat. Technol. 235 (2013) 860–866. [34] H. Wang, R. Zhang, H. Zhang, S. Jiang, H. Liu, M. Sun, S. Jiang, Kinetics and functional effectiveness of nisin loaded antimicrobial packaging film based on chitosan/poly(vinyl alcohol), Carbohydr. Polym. 127 (2015) 64–71. [35] H. Wang, H. Liu, C. Chu, Y. She, S. Jiang, L. Zhai, S. Jiang, X. Li, Diffusion and antibacterial properties of nisin-loaded chitosan/poly(L-lactic acid) towards development of active food packaging film, Food Bioprocess Technol. 8 (2015) 1657–1667. [36] S. Lawrence, D. Adams, D. Bahr, N. Moody, Deformation and fracture of a mudflatcracked laser-fabricated, J. Mater. Sci. 48 (2013) 4050–4058. [37] S. Lawrence, D. Adams, D. Bahr, N. Moody, Environmental resistance of oxide tags fabricated on 304L stainless steel, Surf. Coat. Technol. 285 (2016) 87–97. [38] S. Lawrence, D. Adams, D. Bahr, N. Moody, Mechanical and electromechanical behavior of oxide coatings grown on stainless steel 304L by nanosecond pulsed laser irradiation, Surf. Coat. Technol. 235 (2013) 860–866. [39] H. Espejo, D. Bahr, Substrate cracking in Ti-6Al-4V driven by pulsed laser irradiation, Surf. Coat. Technol. 322 (2017) 46–50. [40] D. Adams, V. Hodges, D. Hirschfeld, M. Rodriguez, J. McDonald, P. Kotula, Nanosecond pulsed laser irradiation of stainless steel 304L: oxide growth and effects, Surf. Coat. Technol. 222 (2013) 1–8. [41] Z. Li, H. Zheng, K. Teh, Y. Liu, G. Lim, H. Seng, N. Yakovlev, Analysis of oxide formation induced by UV laser coloration of stainless steel, Appl. Surf. Sci. 256 (5) (2009) 1582–1588. [42] W. Liu, J. Hansen, Some chemical and physical properties of nisin, a small-protein antibiotic produced by Lactococcus lactis, Appl. Environ. Microbiol. 56 (1990) 2551–2558. [43] L. Coli, V. Todaro, P. Cunha, L. Rodrigues, F. Almada, C. Marques, H. Keiko, H. Castro, C. Rangel, V. Pereira, L. Mendes, Development and characterization of nisin nanoparticles as potential alternative for the recurrent vaginal candidiasis treatment, AAPS PharmSciTech 17 (6) (2016) 1421–1427. [44] M. Iñiguez, M. Avila, E. Iñiguez, P. Guerrero, M. Gutierrez, Antimicrobial activity of disinfectants commonly used in the food industry in Mexico, J. Glob. Antimicrob. Resist. 10 (2017) 143–147. [45] E. Gonzalez, P. Nandy, A. Lucas, V. Hitchins, Ability of cleaning-disinfecting wipes to remove bacteria from medical device surfaces, Am. J. Infect. Control 43 (2015) 1331–1335. [46] G. McDonnell, Antisepsis, Disinfection, and Sterilization: Types, Action, and Resistance, American Society for Microbiology, 2007. [47] V. Suganthi, E. Selvarajan, C. Subathradevi, V. Mohanasrinivasan, Lantibiotic nisin: natural preservative from Lactococcus lactis, Int. Res. J. Pharm. 3 (1) (2012) 13–19. [48] K. Muller-Auffermann, F. Grijalva, F. Jacob, M. Hutzler, Nisin and its usage in breweries: a review and discussion, J. Inst. Brew. 121 (2015) 309–319. [49] Centers for Disease Control and Prevention, [Online]. Available https://www.cdc. gov/infectioncontrol/guidelines/disinfection/disinfection-methods/chemical.html , Accessed date: April 2019.
mesoporous silicas with N-halamine and quaternary ammonium groups for antimicrobial applications, J. Colloid Interface Sci. 533 (2019) 604–611. J. Lin, F. Jiang, J. Wen, W. Lv, Y. Sun, Fluorinated and un-fluorinated N-halamines as antimicrobial and biofilm-controlling additives for polymers, Polymer 68 (2015) 92–100. E. Poverenov, K. Miri, Formation of contact active antimicrobial surfaces by covalent grafting of quaternary ammonium compounds, Colloids Surf. B: Biointerfaces 169 (2018) 195–205. N. Gunther, A. Abdul-Wakeel, E. Reichenberger, S. Al-Khalifa, K. Minbiole, Quaternary ammonium compounds with multiple cationic moieties (multiQACs) provide antimicrobial activity against Campylobacter jejuni, Food Control 94 (2018) 187–194. M. Bravo, G. Preston, R. Van der Hoorn, H. Townley, I. Thompson, Species-specific antimicrobial activity of essential oils and enhancement by encapsulation in mesoporous silica nanoparticles, Ind. Crop. Prod. 122 (2018) 582–590. N. Khorshidian, M. Yousefi, E. Khanniri, A. Mohammad, Potential application of essential oils as antimicrobial preservatives in cheese, Innovative Food Sci. Emerg. Technol. 45 (2018) 62–72. T. Ren, Y. Wang, Q. Yu, M. Li, Synthesis of antimicrobial peptide-grafted graphene oxide nanosheets with high antimicrobial efficacy, Mater. Lett. 235 (2019) 42–45. D. Holdbrook, S. Singh, Y. Choong, J. Petrlova, R. Saravanan, Influence of pH on the activity of thrombin-derived antimicrobial peptides, Biochim. Biophys. Acta Biomembr. 1860 (11) (2018) 2374–2384. L. Juncioni, A. Faustino, P. Gava, T. Vessoni, Nisin biotechnological production and application: a review, Trends Food Sci. Technol. 20 (2009) 146–154. A. Gharsallaoui, N. Oulahal, C. Joly, P. Degraeve, Nisin as a food preservative: part 1: physicochemical properties, antimicrobial activity, and main uses, Crit. Rev. Food Sci. Nutr. 56 (2016) 1262–1274. S. Singh, I. Park, Y. Shin, M. Lee, Y. Lee, Antimicrobial seafood packaging: a review, J. Food Sci. Technol. 53 (6) (2016) 2505–2518. Y. Hung, L. McLandsborough, J. Goddard, L. Bastarrachea, Antimicrobial polymer coatings with efficacy against pathogenic and spoilage microorganisms, LWT 97 (2018) 546–554. H. Yazici, G. Habib, K. Boone, M. Urgen, C. Tamerler, Self-assembling antimicrobial peptides on nanotubular titanium surfaces coated with calcium phosphate for local therapy, Mater. Sci. Eng. C 94 (2019) 333–343. K. Doiron, L. Beaulieu, R. St-Louis, K. Lemarchand, Reduction of bacterial biofilm formation using marine natural antimicrobial peptides, Colloids Surf. B: Biointerfaces 167 (2018) 524–530. A. Héquet, V. Humblot, J. Berjeaud, C. Pradier, Optimized grafting of antimicrobial peptides on stainless steel surface and biofilm resistance tests, Colloids Surf. B: Biointerfaces 84 (2011) 301–309. E. Parfenov, L. Parfenova, G. Dyakonov, K. Danilko, V. Mukaeva, R. Farrakhov, E. Lukina, R. Valiev, Surface functionalization via PEO coating and RGD peptide for nanostructured titanium implants and their in vitro assessment, Surf. Coat. Technol. 357 (2019) 669–683. C. Lee, D. An, H. Park, D. Lee, Wide-spectrum antimicrobial packaging materials incorporating nisin and chitosan in the coating, Packag. Technol. Sci. 16 (3) (2003) 99–106. G. Rossi, J. Han, B. Garcia, E. Castaño, C. Regalado, Effect of temperature, pH and film thickness on nisin release from antimicrobial whey protein isolate edible films, J. Sci. Food Agric. 89 (2009) 2492–2497. J. Chacko, M. Lalpuria, J. Floros and R. Anantheswaran, "Controlled Release of Nisin from Biopolymer Films," The Pennsylvania State University - General Mills
318